Since human cytomegalovirus (HCMV) was first recognized as a threat to the developing fetus, there have been repeated calls for a vaccine that could protect from the damaging effects of HCMV infection in those at risk for HCMV disease. The long quest for a HCMV vaccine that could prevent congenital infection and fetal sequelae, as well as end-organ disease in immune compromised individuals, remains unfulfilled. The primary objective measure for evaluating the efficiency of any vaccine is whether protective levels of immunity are generated and sustained in the vaccinees. An important issue for HCMV is the definition of what constitutes protective immunity. Using a stringent threshold, an immune response can be considered protective only if the vaccinees are absolutely protected from infection following repeated exposure to virus. Alternatively, a vaccine could still be considered protective if the course of challenge virus infection was so dramatically altered that the potential for transmission (horizontal and vertical) and pathogenesis of challenge virus was essentially eliminated. The difference between the two involves the level of virus replication at the primary site of challenge and the extent of dissemination beyond. The former definition requires the generation and maintenance of sterilizing immunity with no spread of the virus. The latter does not, but it does require that the immune system maintain a lifelong restriction on replication of a virus with a complex natural history of persistence in immune competent hosts. The hypothesis is presented that immunization against CMV can generate protective immune responses, although the degree of protection (sterilizing versus limited dissemination) will be dependent on both the titer of challenge virus and the frequency of exposure. According to this hypothesis, immunization can protect completely against infrequent exposure to a low titer CMV challenge. Protection will become more variable as the titer and/or the frequency of exposure to challenge virus increases. Vaccination should shift the virus-host balance decidedly in favor of the host such that both reactivation and shedding are significantly diminished. The hypothesis will be tested in the rhesus macaque model of HCMV infection through the following Aims. (I) Genetic immunization of seronegative macaques with plasmid expression vectors for RhCMV gB, pp65, and IE1, followed by immunization with formalin-inactivated virus. (II) Subcutaneous challenge of vaccinees and controls by experimental inoculation with either high or low titers of RhCMV. (III) Immunization of macaques followed by challenge of vaccinees and controls by natural routes with natural titers of RhCMV by co-housing vaccinees with seropositive, virus-excreting macaques. (IV) Alterations of RhCMV gene expression patterns to induce novel protective immune responses. A CMV vaccine can be considered protective if it results in a dead-end infection. This proposal will stringently test whether a combination of genetic immunization and formalin-inactivated virus can effectively eliminate horizontal spread of RhCMV following either experimental or natural infection.